success (distclean), and almost success (make check).
Dave Korn
dk@artimi.com
Tue Mar 30 17:06:00 GMT 2004
> -----Original Message-----
> From: binutils-owner On Behalf Of Hugh Sasse Staff Elec Eng
> I have tried to replicate the problems that I seemed to have
> with a build then make distclean in the source directory, and
> than a build in a parallel directory. I am pleased to ba
> able to report that this, when done without cvs update -d,
> works successfully.
Hooray! Success at last!
> The straightforward build, just using a parallel directory,
> does fail one test for gas, but I don't know if that is significant.
> Should I be concerned about this?
Not immediately. For a start, that error was also present in your other
build:
--------snip--------
dk@mace /davek> wget
http://www.eng.cse.dmu.ac.uk/~hgs/binutils-cvs-make-distcl
ean-test.txt
--14:49:37--
http://www.eng.cse.dmu.ac.uk/%7Ehgs/binutils-cvs-make-distclean-te
st.txt
=> `binutils-cvs-make-distclean-test.txt'
Resolving www.eng.cse.dmu.ac.uk... 146.227.22.23
Connecting to www.eng.cse.dmu.ac.uk[146.227.22.23]:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 229,489 [text/plain]
100%[====================================>] 229,489 38.26K/s ETA
00:00
14:49:45 (36.01 KB/s) - `binutils-cvs-make-distclean-test.txt' saved
[229489/229
489]
dk@mace /davek> wget
http://www.eng.cse.dmu.ac.uk/~hgs/binutils-cvs-failure.txt
--14:51:18-- http://www.eng.cse.dmu.ac.uk/%7Ehgs/binutils-cvs-failure.txt
=> `binutils-cvs-failure.txt'
Resolving www.eng.cse.dmu.ac.uk... 146.227.22.23
Connecting to www.eng.cse.dmu.ac.uk[146.227.22.23]:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 201,051 [text/plain]
100%[====================================>] 201,051 30.88K/s ETA
00:00
14:51:26 (28.81 KB/s) - `binutils-cvs-failure.txt' saved [201051/201051]
dk@mace /davek> grep ^FAIL: binutils-cvs-*
binutils-cvs-failure.txt:FAIL: CFI on SPARC 64-bit
binutils-cvs-make-distclean-test.txt:FAIL: CFI on SPARC 64-bit
dk@mace /davek>
--------snip--------
So you just overlooked the message. CFI is some kind of dwarf2-related
frame unwinding / exception handling stuff, IIUIC; it's still a bit fluid,
and the occasional bug is likely to show up. If 64-bit SPARC is an
important platform for you, you could always investigate.....
cheers,
DaveK
--
Can't think of a witty .sigline today....
More information about the Binutils
mailing list